Impact of Quantitatively Assessed Interstitial Lung Abnormalities on Long-Term Outcomes After Lung Cancer Surgery

Background: This study evaluated the prognostic significance of quantitatively assessed interstitial lung abnormalities (ILAs) after lung cancer surgery. Methods: We included patients with pathologic stage I non-small-cell lung cancer (NSCLC) who underwent segmentectomy or lobectomy. ILAs were quantified using deep learning texture analysis software. Five-year overall survival (OS) was compared before and after propensity score matching. Competing risks for lung cancer and non-cancer mortality were also analyzed. Results: Among the 1711 patients, 263 (15.4%) comprised the ILA group. The ILA group was older and had a higher proportion of smokers and pathologic stage IB cases (all p < 0.001). The median follow-up period was 48.0 months. Before matching, 5-year OS was significantly worse in the ILA group than in the non-ILA group (82.5% vs. 93.4%, p < 0.001). After 2:1 matching (N = 697), 5-year OS remained lower in the ILA group (85.8% vs. 91.1%, p = 0.025). Multivariable Cox regression analysis showed that the presence of ILAs was associated with increased risk of all-cause mortality (HR 1.52, 95% CI 1.05-2.18, p = 0.025). Restricted cubic spline analysis revealed a nonlinear increase in mortality risk with greater fibrotic ILA burden. In competing risk analysis, death from lung cancer was similar between groups (2.9% vs. 4.2%, p = 0.3), whereas death from other causes was significantly higher in the ILA group (13.0% vs. 3.7%, p < 0.001). Conclusions: Quantitative assessment of ILAs may provide prognostic value in resected stage I NSCLC, particularly in patients with fibrotic changes.